云环境下基于改进遗传算法的虚拟机调度策略  

Virtual machine deployment strategy based on improved genetic algorithm in cloud computing environment

作　　者：袁爱平[1] 万灿军[2] 

机构地区：[1]长沙民政职业技术学院软件学院,长沙410004 [2]湖南大学计算机与通信学院,长沙410082

出　　处：《计算机应用》2014年第2期357-359,364,共4页journal of Computer Applications

基　　金：湖南省教育规划基金资助项目(XJK011BXJ006);湖南省科技计划项目(2012GK3062)

摘　　要：针对云环境下服务器内部多种资源间分配不均衡问题,提出了一种多维资源协同聚合的虚拟机调度算法MCCA。该算法在分组遗传算法的基础上,采用模糊逻辑及基于资源利用率多维方差的控制参量,设计适应度函数指导搜索解空间。算法使用基于轮盘赌法的选择方法,并对交叉和变异等进行了优化,以实现快速有效地获取近似最优解。在CloudSim环境下进行了仿真,实验结果表明该算法对均衡多维资源分配和提高资源综合利用率具有一定的优势。Aiming at improving the resource utilization of data center by balanced usage of multiple resources, a scheduling algorithm based on group genetic algorithm for multi-dimensional resources coordination was proposed to solve the virtual machine deployment problem. To guide the solution searching, a fuzzy logic based multi-dimensional fitness function was raised. Meanwhile, innovative optimization of crossover and mutation was put forward to improve the solution quality. The results of simulation in CloudSim environment prove that using the proposed algorithm can obtain better multi-dimensional resources performance and higher resource utilization rate.

关 键 词：云计算 虚拟机 多维均衡 分组遗传算法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]